JazzHR Jobs API | Search Every Company Job Board
Pricing
from $1.00 / 1,000 job records
Search live job postings across every company hiring on JazzHR (applytojob.com). No company URL needed. Returns titles, employers, locations, posting dates, apply links, and full descriptions in markdown, HTML, or text, plus a directory of every employer on the platform.
Optional. JazzHR company boards such as https://company.applytojob.com or individual job URLs. Leave empty to search every company on the platform.
Optional. Keep only jobs whose title contains any of these words, for example nurse, welder, account executive. Matched against the job URL before each job page is fetched, so filtering here makes runs cheaper.
jobs returns full job records with descriptions. urlsOnly returns the job index without opening each posting, which is the cheapest way to list everything. companiesOnly returns one row per employer with a job count, a directory of every company hiring on JazzHR.
How to return each job description. Markdown is compact and reads well in AI agents and LLM pipelines, html preserves the original markup, text is plain prose, all returns every format.
Optional. Company slugs to include, taken from the subdomain of a JazzHR board. For https://acmecorp.applytojob.com the slug is acmecorp.
Optional. Company slugs to skip. Useful for filtering out staffing agencies or employers you already track.
Optional. Case-insensitive text match against the city, state, postal code, or country of the job, for example Austin, CA, or 92618.
Keep only jobs of this employment type. Leave on Any to include everything.
Keep only postings the employer flagged as remote.
Optional. Keep only jobs the employer posted on or after this date, in YYYY-MM-DD form. Uses the real posting date from the job page.
Stop after this many rows. The platform holds well over 100,000 live jobs, so this cap is what keeps an exploratory run small and cheap.
Return only postings this Actor has not returned in an earlier run. Job IDs are remembered between runs, which turns a schedule into a feed of newly posted jobs.
What the first run should do when Only jobs not seen before is on. Emit all returns everything and remembers it. Seed only returns nothing and just records what exists today, so the next run is a clean list of new jobs.
Named key-value store that holds the job IDs seen so far. Change it to keep separate delta histories for separate searches.
How many job pages to read at once. Lower it if the source starts rejecting requests; the Actor also slows itself down automatically when error rates rise.
